Free short course
Prove It
Catching AI when it is wrong, before it costs you. Four modules. Five checks. A habit that keeps your name off the mistake.
This course is free in full: every module, knowledge check and the final task, with no sign-up.
AI tools are confident. They are confident when they are right and exactly as confident when they are wrong, and the wrong answers are written just as fluently as the right ones. If your job breaks when you publish something false, and most jobs do, that is a problem you need a method for.
This course gives you the method. It explains, without jargon, why AI tools invent things. It gives you five checks that catch most errors in under five minutes. It shows you how to ask for sources in a way that produces real ones. And it builds the habit of leaving a trail, so that when someone asks "did you check that," the answer is yes, and here is how.
It is for people who use AI for work that other people rely on: reports, advice, figures, summaries, anything with your name on it.
By the end, you'll be able to…
- Explain in plain English why an AI tool produces confident, fluent falsehoods, and which kinds of question are most at risk.
- Run the five-check routine on any piece of AI output in under five minutes.
- Get real, checkable sources from an AI tool, and recognise a fabricated citation on sight.
- Decide in seconds whether a task is low, medium or high stakes, and match the checking effort to it.
- Keep a verification trail that would satisfy a manager, an auditor or a client.
What's in it
- Module 1: Why it lies
It does not know it is wrong, and neither will you unless you look.
- Module 2: The five checks
A routine that takes five minutes and catches most of what matters.
- Module 3: Real sources
Getting references that exist, and spotting the ones that do not.
- Module 4: Stakes and trail
Match the effort to the risk, and leave evidence that you checked.
- Low, medium, high 6 min
- The trail 7 min
- When you find it wrong 6 min
- Module 4 knowledge check
- Finishing the course
One final task that puts the whole method to work on something real.